Re: [hpoj-devel] scan lockups with PSC500

[<pa...@rc...>]

dgun said:
> OK.  Here's some debug output right before the timeout.  If you want
> the entire debug output, let me know.  This one didn't even get past
> the initial negotiation: 
...
> Oct 17 21:14:02 localhost kernel: status 1: 5f
> Oct 17 21:14:02 localhost kernel: c60b11c0: event=9.
> Oct 17 21:14:02 localhost kernel: status 2: 2f
> Oct 17 21:14:02 localhost kernel: get_nibble(c60b11c0): ni4>  state=5
> event=9 reset=7 recv=1 rcv=0 send=2
> Oct 17 21:14:02 localhost kernel: status 2: 6f
> Oct 17 21:14:02 localhost kernel: mlcpp0: mlcpp_intr c60b11c0
> Oct 17 21:14:02 localhost kernel:   state=5 event=9 reset=7 recv=1 rcv=0
> send=2

Hi, Daniel.  Unfortunately, it still got corrupted here, which might explain
why it appeared to not get past the initial negotiation.  Please try the
attached patch against a clean copy of ieee12844pp.c (version 0.6).  I yanked
out all but the most interesting log messages in hopes of reducing the
gaps.  However, I can't seem to actually reproduce the problem (on an R80,
which is of similar vintage to the PSC500), where I'm running
"ptal-connect mlc:mlcpp0 -scan" and typing <esc>E and <esc>*f0S.  Perhaps
there is a slight timing difference between my computer and yours, and/or
between the R80 and PSC500, which is causing this problem for you but not
for me.  I'll try to experiment with a PSC500 at work sometime soon.

David